Drug shortages continue to affect the health system in Canada. They can leave patients and caregivers struggling to fill prescriptions or to find medications locally, and they put additional stress on healthcare providers, hospitals, and clinics.

A new platform called MaaTRx, developed by researchers at the University of Toronto, aims to provide healthcare leaders with a clearer view of which drugs are at risk for shortage, allowing them to take steps earlier to reduce the impact on patients and the health system.

The platform uses real-time data and AI-enabled forecasting to identify medications that may be vulnerable to supply disruptions. This information enables policymakers, pharmacy networks, and hospitals to make better-informed decisions about their inventory and look for alternative products sooner.

“Drug shortages are a huge and persistent issue in Canada,” Mina Tadrous, PharmD, professor at the Leslie Dan Faculty of Pharmacy at the University of Toronto in Toronto and cofounder and scientific lead for MaaTRx, told Medscape News Canada. “Pharmacists and doctors are likely dealing with this on a weekly basis.”

The Health Products Shortages Canada website currently lists 1761 shortages, with 123 drugs anticipated to be in shortage soon. Data for this Health Canada-funded website come directly from manufacturers.

The concept behind MaaTRx builds on research led by Tadrous that helped inform Health Canada’s development of Canada’s Critical and Vulnerable Drug List.

Broad Impact of Drug Shortages

“Right now, shortage management is almost entirely reactive,” Christine Leong, PharmD, an associate professor in the College of Pharmacy at the University of Manitoba in Winnipeg, told Medscape News Canada. “In the hospital setting, when shortages come up, [pharmacists] can be left scrambling to figure out what the best alternative is.”

Also, “even though drug manufacturers are required to report shortages, those reports often arrive late,” said Leong. “Sometimes, community pharmacists don’t find out that a drug isn’t available until the wholesale order doesn’t show up.”

This belated communication can raise concerns for patient safety and clinical outcomes. “This is why we developed the solution that we did,” said Tadrous, “because some shortages can lead to increased mortality, like when a hospital doesn’t have enough cancer drugs.”

Shortages also require pharmacists and other health professionals to spend extra time figuring out the best alternative drugs for patients. “That’s quite a few hours of duplicated work during every major shortage,” said Leong. “So having a tool like this, which gives us some lead time and shared visibility into drug shortages, could help address a real gap.”

Nevertheless, Leong highlighted that this tool can’t address the root causes of drug shortages, which may include manufacturing issues, difficulty obtaining raw materials or ingredients, sudden increases in demand, discontinuation of products, or natural disasters and pandemics.

Still, “the tool could help with coordination, especially with critical drugs,” she said. “By knowing weeks in advance which drugs are at risk of a shortage, we can better plan therapeutic substitutions,” she said.

Real Data Couple With AI Forecasting

One problem with drug shortage reports from manufacturers, according to Tadrous, is that they “often don’t lead to an actual shortage, because the system has other [similar] drugs. So these reports send a lot of false signals, with people overreacting or underreacting.”

“If there are too many false positives,” said Leong, “clinicians could get alert fatigue and start to tune out the alerts.”

To improve the accuracy of predictions, “MaaTRx uses real-world purchasing data and links together dozens of data sources to better understand the drugs’ availability,” said Tadrous.

The platform examines “the likelihood that the drug will go into a real shortage in the country during the next year,” he said, “which provides a runway to plan ahead.” This prediction is updated dynamically each month based on real marketplace data.

“We then take that data and feed it through AI, using algorithms that we’ve built, to give clinicians on the ground the ability to better plan when a shortage happens,” said Tadrous.

Planning at Hospital and Regional Level

MaaTRx acts like an early warning system, allowing health teams to plan and avoid having to scramble once a medication disappears from shelves. This platform supports decision-making at hospitals and clinics, as well as at the regional or national level.

Testing showed that of the drugs that MaaTRx flagged as being at high risk for a shortage during a validation period, 25% went into a real shortage during that time, said Tadrous. However, “just like any other risk score, not everything goes into a shortage,” he said. “Some of these drugs, though, may end up going into a shortage later on, just not within the next year.”

Overall, “we’re really proud of this platform, because we’re putting it in the hands of people who can make the decisions, and we’re teaching people how to make better decisions for drugs that matter,” he said.

Tadrous reported being cofounder and scientific lead for healthcare data company MaaTRx. Leong reported no relevant financial relationships.